Data Analytics & Research
Reporting results properly: p-values, effect sizes and confidence intervals
GOSPELTRADER Research Desk · 30 September 2026 · 7 min read
Quick answer
Report three things for every key result: the estimate (the effect size), its 95% confidence interval, and the exact p-value. The estimate says how big, the interval says how precise, and the p-value says how surprising the data would be if there were no effect.
"The result was significant (p < 0.05)" is the most common sentence in Nigerian project chapters, and on its own it says very little. With a large enough sample, a trivially small difference becomes significant.
A better template
Instead of "income significantly affected savings", write: "Each additional ₦50,000 of monthly income was associated with ₦6,200 more in monthly savings (95% CI ₦3,900 to ₦8,500; p < 0.001)." A reader now knows the direction, the size, the uncertainty and the evidence strength.
Common effect sizes
| Test | Effect size | Rough 'small / medium / large' |
|---|---|---|
| t-test | Cohen's d | 0.2 / 0.5 / 0.8 |
| ANOVA | Eta squared (η²) | 0.01 / 0.06 / 0.14 |
| Correlation | r | 0.1 / 0.3 / 0.5 |
| Chi-square | Cramér's V | depends on table size |
Non-significant is still a result
A wide interval that includes zero means the study could not tell. Say so plainly, report the interval, and discuss sample size rather than calling the hypothesis 'rejected'.